vix.ing · top · new · best · stats · spec

A Baseline Multimodal Approach to Emotion Recognition in Conversations

2026/01/31 by Víctor Yeste, Rodrigo Rivas-Arévalo · 1 voice
Computer Science · Engineering · #cs.AI #cs.CL #cs.CY #cs.SD #eess.AS

paper · pdf · doi:10.48550/arxiv.2602.00914

Abstract

We present a lightweight multimodal baseline for emotion recognition in conversations using the SemEval-2024 Task 3 dataset built from the sitcom Friends. The goal of this report is not to propose a novel state-of-the-art method, but to document an accessible reference implementation that combines (i) a transformer-based text classifier and (ii) a self-supervised speech representation model, with a simple late-fusion ensemble. We report the baseline setup and empirical results obtained under a limited training protocol, highlighting when multimodal fusion improves over unimodal models. This preprint is provided for transparency and to support future, more rigorous comparisons.

Citations

Discussions

Related